The goal of this project is to develop a prognostic test for identifying early stage breast cancer (BC) patients with biologically agressive disease who are at high risk for relapse despite adjuvant therapy. This will be accomplished by evaluating patients for occult micrometastases at diagnosis and correlating the invasive nature of these cells with clinical outcome. The short term objectives are to determine the feasibility of an in vitro micrometastatic residual cancer invasion (MRCI) assay. For development of an in vitro invasion model for breast cancer, tumor cells will first be seeded in normal bone marrow and plated on an artificial basement membrane in tissue culture wells containing medium. Tumor cells are incubated on the membrane for a period of 72 hours. The cells that invaded the artificial basement membrane are then fixed, stained by tumor specific antibodies and enumerated. A pre-clinical model will be developed first using breast cancer cell lines, then this model will be tested on stage IV BC patients, and finally on early stage BC patients. If successful, this assay will provide a simple, reliable and fast clinical test that will assist the treating physician in therapeutic decision making. PROPOSED COMMERCIAL APPLICATION In 1997 approximately 160,000 women will be diagnosed with early stage Breast Cancer. Approximately 50,000 women will have occult tumor cells in their marrow at diagnosis, of which 25,000 will not relapse. The Minimal Residual Cancer Invasion (MRCI) test will identify biologically agressive marrow micrometastases and allow clinicians to identify high risk patients at diagnosis.